Studies on ductility of RC beams in flexure and size effect

G. Appa Rao* & I. Vijayanand

*University of Stuttgart, 70569, Stuttgart, Germany
Indian Institute of Technology Madras, Chennai-600 036, India
R. Eligehausen
*University of Stuttgart, 70569, Stuttgart, Germany

ABSTRACT: This paper reports on some experimental investigations on ductility of reinforced concrete
beams in flexure and evaluation of size effect. The minimum flexural reinforcement has been evaluated from
experimental observations on ductility of reinforced concrete (RC) beams. Beams of depth 100mm, 200mm,
400mm at different flexural reinforcements namely 0.15, 0.30, 0.60 and 1.0 % were tested under uniform
bending moment. The beams were made of 30 MPa concrete. The cracking and ultimate flexural strength, in-
fluence of beam depth on ductility and rotation capacity have been analyzed. The size of RC members has a
significant influence on flexural behaviour. The variation of cracking strength is not very conclusive in small-
size beams, where as it decreases as depth of beam increases beyond 200 mm. The ultimate flexural strength
has been observed to decrease as the beam size increases. As the flexural reinforcement ratio in beams in-
creased the ductility of beams was observed to increase. Ductility of RC members decreases with increase of
beam size. The optimum flexural reinforcement has been obtained from an optimum ductility number, Np,
equal to 0.20. The minimum flexural reinforcement was observed to decrease as the beam depth increased,
and decreased as the yield strength of steel reinforcement increased.

1 INTRODUCTION beams subjected to flexure, shear and torsion, asso-

The expressions for minimum steel ratios, both in
flexure and shear, prescribed by various codes of
practice are basically empirical. Due to lack of ra-
tional approach, the code provisions for design of
RC members show vast variation in the design val-
ues. A minimum area of reinforcement is generally
required in flexural members to prevent cracking
and excessive deflections due to various loading ef-
fects. Two parameters such as tensile strength of
concrete and yield strength of reinforcement are in-
corporated in the expressions for predicting mini-
mum flexural reinforcement in RC beams. However,
the effect of size of RC members is not considered.
In some lightly reinforced beams, the cracking mo-
ment, as a plain concrete beam, may exceed the
yielding moment of the beam, as an RC beam, at
first cracking. This criterion has been considered for
evaluating the minimum flexural steel ratio by ACI
code (ACI-318-2005). The Indian standard (IS: 456-
2000) specifies minimum flexural reinforcement to
avoid sudden failures of RC members based on sim-
ple assumption that the yielding moment of RC
beam, My is greater than or equal to cracking mo-
ment, Mcr, of plain concrete beams.
Recently, a theoretical evaluation of minimum
longitudinal and transverse reinforcement ratios in
ciated with ductility and minimum strength at ulti-
mate limit state (ULS) has been reported (Shehata et
al., 2003). The behavior of RC beams is useful to
postulate a rational approach to estimate minimum
flexural reinforcement based on rational approach.
Carpenteri (1984) defined a brittleness number to
determine the minimum flexural reinforcement in
RC beams using fracture mechanics principles. The
brittleness number is defined as “Np”
1
2
fyh ⎡ As ⎤
N p = ⎢ A ⎥ (1)
K Ic ⎣ ⎦
Where fy = yield strength of steel reinforcement, KIc
= concrete fracture toughness, As = area of steel re-
inforcement, A = area of c/s of beam, h = beam
depth
The brittleness of structural member increases as
its size increases and/or the reinforcement ratio de-
creases. Physically similar behaviour was revealed
in some cases where the brittleness number Np was
the same. At a value of Np equal to 0.26 using 91
MPa concrete, the yielding moment was more or less
equal to the first cracking moment of the beam. The
reinforcement corresponding to this condition was
considered for evaluating the minimum reinforce-
ment for flexural members. The minimum percent-
age reinforcement tends to be inversely proportional
to the beam depth, while the values by codes are in-
dependent of the beam depth. Strain localization has
been taken into account for the analysis of RC
beams by Hillerborg (1990). The descending portion
occurs due to crack formation within fracture proc-
ess zone. The analysis of RC beams for balanced re-
inforcement ratio decreases with increasing beam
depth.
Bosco et al. (1990) reported that the minimum
flexural reinforcement corresponds to a condition at
which the formation of first flexural cracking and
yielding of steel reinforcement occur simultane-
ously. The brittleness of RC beams increases as the
size of beam increases and as steel ratio decreases.
An optimum value of Np has been observed for esti-
mating the minimum flexural reinforcement corre-
sponding to the above condition. The minimum steel
ratio is inversely proportional to depth of beam.
Bosco et al. (1990) reported that the brittleness of
beams increases by increasing size-scale and/or de-
creasing steel area. For low Np values in lightly rein-
forced beams or for small cross sections the fracture
moment decreases while the crack extends. The peak
or first cracking load is lower than the steel-yielding
load only at high brittleness number. The size of
beam seems to govern the post peak behavior, espe-
cially for low brittleness number for larger beam
depth.
Baluch et al. (1992) proposed a criterion for
minimum flexural reinforcement with which unsta-
ble crack propagation was avoided. This is achieved
by ensuring that the moment corresponding to the
maximum load, in a reinforced concrete beam is
greater than its cracking moment as a plain concrete
beam. The expression proposed to predict the mini-
mum flexural reinforcement is
1.9134 K Ic 0.82
ρ min = (2)
f y 0.9922 (1.7 − 2.6 c s / D )
Gerstle et al. (1992) used fictitious crack model
to study tensile cracking behaviour of singly rein-
forced concrete beams in flexure. A theoretical
analysis was performed to plot normalised moment
with the normalised crack length for different values
of “β” (a measure of brittleness) and “α” (a measure
of steel percentages). Stable crack propagation has
been associated with a continuously increasing curve
and that value of “α” corresponds to minimum flex-
ural reinforcement. An expression for minimum re-
inforcement ratio has been proposed defined but it
does not contain fy as shown below.
E ⎛ ft D ⎞
1/ 2
(3)
ρ min = C ⎜⎜ 0.0081 + 0.0148 − 0.0900 ⎟
⎟
ES ⎝ E c wc ⎠
2 RESEARCH SIGNIFICANCE
The design of members based on strength criteria
does not consider fracture mechanics theory. Fail-
ures in RC members exhibit different modes due to
change of beam depth and percentage reinforcement.
The ductility of RC members changes with size of
member and strength of concrete. However, this is
not considered in the design of structural members.
In other words, failure according to strength theory
should not show any size dependence, nor should
the size of beam have any effect on its ductility. The
effect of size of member and ductility in design of
RC members can be predicted by fracture mechan-
ics. However, there exists a controversy in the
evaluation of minimum flexural reinforcement in RC
members. An attempt has been made to understand
size effect on ductility and minimum reinforcement
of lightly reinforced beams.
3 EXPERIMENTAL PROGRAMME
3.1 Materials
An Ordinary Portland Cement (OPC) was used for
the present study. The properties of cement are pre-
sented in Table 1. The fine aggregate was obtained
from a natural river bed. The aggregate fraction
passing through sieve size 1.18 mm and retained on
600 μ size was used in concrete. The specific gravity
and fineness modulus of sand are given in Table 2.
The machine crushed granite aggregate was used for
concreting, consisting of mixture of 10 and 20mm
size particles. The properties of aggregate are given
in Table 3. Potable water was used for mixing of
concrete and curing of specimens. The pH value of
water was 7.8.
Table 1. Properties of cement
S.No. Property Results
1 Compressive 43.0
Strength N/mm2
3 Fineness 3.5
4 Initial setting time 205 min
5 Final setting time 335 min
6 Specific gravity 3.12
The flexural reinforcement was high strength
steel reinforcement with 415 N/mm2 guaranteed
yield strength. The diameter of the bars varied from
3 to 12mm depending on the size of beam. The
nominal shear reinforcement consists of MS bars of
diameter from 3mm to 6mm depending on the size strengths of 6mm, 10mm, and 12mm diameter bars
of beam. were 577, 483 and 459 N/mm2 respectively. The
ductility number defined by Carpenteri (1984) was
Table 2. Properties of fine aggregate used for evaluation of minimum flexural reinforce-
S.N Property Result
ment. Fracture energy, GF of plain concrete was de-
o
termined on three-point bend specimens (depth, d =
1 Specific gravity 2.78
100mm, width, b = 100mm, and span, l = 500 mm).
2 Fineness modulus 2.82
The notch-to-depth ratio was 0.5 and the notch
width was 3mm. The mean value of fracture energy
was 150 N/m. The critical stress intensity factor was
Table 3. Properties of coarse aggregate evaluated as K1C= √GF √E and was equal to 64.09
S. No Property Results N/mm3/2. The modulus of elasticity of concrete was
1 Specific gravity 2.70 27.40GPa.
2 Fineness Modulus 6.84 A total of 20 RC beams with depth equal to 100,
200mm and 400mm, maintaining the depth-to-width
ratio 2.0 were tested. For a particular parameter, two
3.2 Specimen details beams in class A and B and only one beam in class
C were tested. The span between supports was equal
Rectangular beam specimens of different depths
to five times beam depth, d. Therefore, the spans
were adopted. In order to maintain geometric simi-
measured 500, 1000 and 2000 mm respectively in
larity, the aggregate size was varied depending on
beams (for class A: d = 100 mm, b = 50 mm; l =
the depth. In small beams of size 50mm x 100mm x
500; class B: d = 200 mm, b=100 mm; l=1000 mm;
500mm and 100mm x 200mm x 1000mm, 10mm
and class C: d = 400 mm, b = 200 mm; l = 2000
aggregate was used, while 20mm aggregate was
mm). The reinforcement was estimated from ductil-
used in large beams of size 200mm × 400mm ×
ity numbers selected i.e. 0.091, 0.183, 0.366 and
2000mm. The ratio of reinforcement cover-to-depth
0.732. Steel moulds were used for casting the beams
was 0.05 in all the beams.
of required dimensions.
Table 4: Beam dimensions and reinforcement details.
3.3 Experimental set-up and testing
Beam σy, Ast Provided Stirrups Four-point loading set-up was used for testing of RC
MPa (%) mm2 No. of bars Spacing, mm beams as shown in Figure 1. Statically determinate
A1 637 (0.30) 14 2–3mm 3mm @150 system was ensured by adopting hinge and roller
B1 637 (0.15) 28 4–3mm 3mm @ 300
C1 389 (0.15) 113 4-6 mm 6mm @ 150 supports at two ends. The load was applied through
A2 637 (0.3) 14 2–3mm 3mm @150 a hydraulic jack at constant load increments. The
B2 389 (0.3) 56.5 2–6mm 6mm @ 140 load was applied symmetrically at one third points.
C2 459 (0.3) 226 2-12 mm 6mm @ 300 LVDT was used to measure the deflection at the
A3 637 (0.6) 28 4–3mm 3mm @150
B3 389 (0.59) 113 4–6mm 6mm @ 140 center of beams.
C3 459 (0.59) 452 4-12 mm 6mm @ 130
A4 389 (1.19) 56.5 2–6mm 6mm @ 70
B4 577 (1.0) 756 6-12mm 8mm @ 175
+1-10mm
C4 459 (1.0) 756 6-12 mm 8mm @ 175
+1-10mm

The details of beam specimen and steel reinforce-

ment are given in Table 4. The concrete mix propor-
tions were 1: 2.75: 5.1 respectively cement: fine ag-
gregate: coarse aggregate. The cement content was
250 kg/m3 and the water cement ratio was 0.75. The
compressive strength of concrete at 28 days on
100mm size cubes was 30 N/mm2. The split tensile
strength of 150mm x 300mm cylinders was 2.62
N/mm2. The beam as well as companion cube and
cylindrical specimens were cured in water for 28
days. The specimens were tested after 28 days. The
steel reinforcement consisted of 3mm, 6mm, 10mm,
and 12mm diameter bars as flexural reinforcement.
The actual yield strength of reinforcement was used
to calculate the flexural strength of beams. The yield
Figure 1. Experimental Set-up.
The load at first cracking was visualized by means
of magnifying glass in the uniform bending moment
region between two central loading points where the
first flexural crack was formed. The strains along the
depth of the beam were measured using demount-
able mechanical gage. The crack propagation was
monitored on the beam surface and the crack width
was measured by a microscope. The ultimate mo-
ments of the beam were estimated both theoretically
and experimentally.
4 RESULTS AND DISCUSSION
4.1 Load-deflections curves
It was noticed that the deflection of beams at failure
increased as the percentage steel reinforcement in-
creased from 0.15 % to 1.0 %. It was observed that
the ductility of beams increased as the flexural rein-
forcement increased. The beam with 0.30 % rein-
forcement exhibited large deflection at failure show-
ing increased rotation capacity and ductility. Similar
trend was observed in all the beams i.e. deflections
increase with percentage flexural reinforcement
keeping other parameters constant. At 0.30 % rein-
forcement the beams exhibited improved ductility.
Interestingly, the nature of failure changed from
ductile to brittle as the depth of beam increased. The
large size beams exhibited relatively small deflec-
tion at failure. As the size of beam increased, keep-
ing the percentage flexural reinforcement constant,
the deflections at failure decreased. This shows that
as the beam size increases the failure of beams
turned from ductile to brittle. The ductility of beams
was found to decrease with increase of depth. Fur-
ther it was observed that as the percentage rein-
forcement increased at a given beam depth the duc-
tility increased.
4.2 Flexural strength
The flexural strength is defined as the flexural ca-
pacity of the beam at the ultimate load. In this case,
nominal strength of beam at the ultimate load was
represented by its flexural strength. The nominal
strength was calculated by dividing the ultimate load
by square of depth of beams in three dimensional
similarities.
0.18 p = 0.15
p = 0.60
0.16
0.14
0.12
2
Mu/fck bd
0.1
0.08
0.06
0.04
0.02
0
0 100 200 300 400
Depth, mm
Figure 2. Moment ratio vs. Depth at different steel ratio.
Figure 2 shows the variation of flexural strength
calculated as (Mu/fck bd2) with depth of beam at dif-
ferent flexural reinforcements i.e. 0.15, 0.30, 0.60
and 1.0%. The flexural strength has been observed
to decrease with depth at 0.15 and 0.30 % rein-
forcements. However, the nondimensional flexural
strength has been observed to decrease as the beam
depth increased at 0.60 and 1.0 % reinforcements.
This shows that a general size effect law has been
possible in R.C. beams in flexure at small percent-
age reinforcements. However, at heavy flexural rein-
forcements, the effect of size needs further studies.
At this juncture it would not be possible to conclude
the exact size effect on flexural strength of rein-
forced concrete beams with heavy reinforcement.
4.3 Ductility factor
Ductility factor may be defined as the ratio of de-
flection at failure to the deflection at yield or at the
first crack. As there is no information on the effect
of size on ductility of reinforced concrete beams, the
present study was undertaken. In codes of practice,
the design strength of RC members in flexure is con-
sidered to be constant. When the concepts of fracture
mechanics are used, there could be an improved
safety margin against failure and the prediction of
failure could be possible with reasonable reliability.
Figure 3 shows variation of ductility factor with size
of structure at 0.15, 0.30, 0.60 and 1.0 % reinforce-
ment. It demonstrates that the ductility factor in-
creases as the beam size increases from 100mm to
200mm. Thereafter, the ductility factor decreases
with size.
At small flexural reinforcement ratios, the ductil-
ity factor has been observed to be the highest at
200mmbeam depth. At higher percentages of rein-
forcement, the trend seems to be increasing with size
of structure. At 100mm depth, the ductility at all
percentages of reinforcement was found to be the
lowest. However, as the depth of the beam increases
beyond 200mm, the ductility factor has been show-
ing size dependence. Further at small percentage of
reinforcement, the ductility factor increases with in-
creasing percentage reinforcement.
p =0.30
p = 1.0
0.15 0.3
9
0.6 1.0
8
7
Ductility Factor
6
5
4
3
2
1
500
0
50 100 150 200 250 300 350 400 450
Depth, mm
Figure 3. Ductility factor vs. beam depth.
4.4 Ductility number
The ductility number is a measure of brittleness of
R.C. beams. The study was designed to investigate
the effect of beam size and percentage reinforcement
on ductility of R.C. beams. In order to evaluate the
effect of beam size on ductility three different beam
sizes i.e. 100mm, 200mm and 400mm were adopted
by varying the flexural reinforcement at 0.15, 0.30,
0.60 and 1.0 %. Figure 4 shows the ductility number
with beam depth at all percentages of reinforcement.
It demonstrates that the ductility number increases
with size of beam at a given percentage flexural re-
inforcement. The ductility number has been found to
increase as the beam depth increased. Similar trend
has been observed in the case of beams reinforced
with 0.6 and 1.0 % flexural reinforcement. It was
observed from the load-deflection curves that the
beams turned brittle with increase in depth. The in-
crease of the ductility number at 0.6 and 1.0 % was
significantly higher at a given size of beams. The
ductility number for 200mm deep beams was 0.198,
beyond which it was found to decrease with increase
in beam depth. This value of 0.198 is the optimum
value for achieving minimum required ductility for
evaluation of the minimum reinforcement.
1.6
1.4
1.2
1
Ductility Number
0.8
0.6
0.4
0.2
0
50 100 150 200 250
Depth, mm
Figure 4. Ductility number vs. beam depth.
0.6
0.5
0.4
Min. % Reinforcement
0.3
0.2
0.1
0
50 250 450 650 850
Depth, mm
Figure 5. Min reinforcement vs. beam depth.
Figure 5 shows the minimum flexural reinforce-
ment with beam depth at the optimum ductility,
which corresponds to the value of Np equal to 0.198
≈ 0.20. As the strength of concrete increases the brit-
tleness of beam increases due to material brittleness.
Similar observations have been made in high
strength concrete beams by Bosco et al. (1990),
where the ductility number for achieving the mini-
mum flexural reinforcement was 0.26. Therefore, in
order to maintain the minimum ductility in RC
members, the percentage reinforcement should be a
function of strength of concrete, depth of beam and
yield strength of steel reinforcement.
5 CONCLUSIONS
The following conclusions were drawn from the
studies on lightly reinforced concrete beams.
1. The effect of size and percentage reinforcement
on the ultimate strength of RC beams has been
found to be significant. The ultimate strength is
inversely proportional to the beam depth.
2. As the percentage flexural reinforcement in-
creases, the ultimate load and the corresponding
the beam deflections increase. As the depth of
beam increases the ductility factor decreases.
3. The ductility number of RC beams increases
with increasing beam depth and with decreasing
0.15 0.3
percentage reinforcement. The optimum ductility
0.6 1.0 number is 0.20 in 30 MPa concrete.
4. The minimum percentage reinforcement is in-
versely proportional to beam depth. It indicates
that the formula for minimum steel reinforcement
provided by the codes needs to be modified.
300 350 400 450
REFERENCES
ACI 318-2005. Building code requirements for structural con-
crete and commentary, ACI 1995 Farmington Hills, MI:
fy = 439 MPa Baluch, M. H. Azad, A. K. and Ashmawi, W. 1992. Fracture
fy = 550 MPa
mechanics application to reinforced concrete members in
fy = 250 MPa
flexure. Proc. Int Workshop on application of fracture me-
chanics to reinforced concrete, Italy, 413-436.
Bosco, C. Carpinteri, A. Debernardi, P.G. 1990. Minimum re-
inforcement in high strength concrete. Journal of Structural
Engg 116(2): 427 – 437.
Bosco, C. Carpenteri, A. and Debernardi, P.G. 1990. Fracture
of reinforced concrete; scale effects and snap – back insta-
1050 1250 1450 1650
bility. Engg. Frac Mech 35(4/5): 665 - 677.
Carpinteri, A. 1984. Stability of fracturing process in RC
beams, ASCE Jl of Str Engg, 110 (3), 544-558
Gerstle, W. H. Dey, P. P. Prasad, N. N. V. Rahulkumar, P. and
Xie, M. 1992. Crack growth in flexural members-A frac-
ture mechanics approach. ACI Str Jl, 89(6): 617-625.
Hillerborg, A. 1990. Fracture mechanics concepts applied to
moment capacity and rotational capacity of reinforced con-
crete beams. Eng. Fra. Mech 35(1/2/3): 233-240.
IS 456-2000. Code of practice for design of plain and rein-
forced concrete structures, BIS, New Delhi.
Shehata, I.A.E.M. Shehata, L.C.D. and Garcia, S.L.G. 2003.
Minimum steel ratio in RC beams made of concrete with
different strength–theoretical approach. Mat Str 36: 03-11.